Add comprehensive unit tests for the xecuteConditions rule engine in
packages/shared/src/utils/conditions.ts, which previously had no test
coverage. Covers all 12 comparison/existence rules, logical AND/OR/NOT
nesting, implicit AND across multiple conditions, and dot-path
resolution (including missing-path and whole-value fallback). Uses the
existing vitest setup in the shared package.
- Load the ext-apps bridge lazily. AppBridge/PostMessageTransport now come from a
dynamic import() inside Effect 1, with a type-only import at the top. A
<CopilotKit> app no longer pays the ~40-50 kB gzipped ext-apps cost unless it
actually renders an MCP App. Verified: the built output has no static ext-apps
import, only import("@modelcontextprotocol/ext-apps/app-bridge").
- Move @modelcontextprotocol/sdk out of dependencies. It is now an optional
peerDependency (mirroring how ext-apps declares it) plus a devDependency for
our own build, so the sdk tree (express, hono, jose, ajv, cross-spawn) is no
longer an install/audit surface for every React consumer. Nothing in the
bundle reaches sdk at runtime; the only sdk usage is a type import.
- Raise the zod peer floor to >=3.25. The ext-apps/sdk schema slice imports
zod/v4 and zod/v4-mini, which only exist in zod >= 3.25; the old >=3.0.0 peer
let a consumer on zod 3.24 hit an unresolvable import at build time.
- Seed the host context at AppBridge construction (hostContext option) instead of
calling setHostContext after connect, so it is deterministically in place when
the widget's ui/initialize is handled (the seam #6689 needs to advertise
displayMode / availableDisplayModes at initialize).
- Restore the full cross-frontend testid surface-contract comment.
- Split the oncalltool guard so "no server hash" and "no agent" report distinctly.
Angular had no counterpart to `useComponent` in react-core and vue, so the only
way to let an agent display one of your components was `registerFrontendTool`
with a handler the component does not need, or `registerRenderToolCall`, which
requires the tool to already exist in the agent.
`FrontendToolConfig.handler` becomes optional, and `#bindClientTool` returns the
config unwrapped when it is absent. Core already has the render-only path: a tool
declaring no handler gets an empty tool result and completes the turn. Binding a
wrapper regardless called `undefined` on the first tool call, and a stub handler
would write a fabricated result into the thread instead.
`registerComponent` is that shape with the model-facing description built for it,
identical to the one react-core and vue build, so the tool reads the same to the
model whichever frontend registered the component. The component stays an
ordinary `ToolRenderer` reading `toolCall().args`.
The tool is declared by the frontend and forwarded over AG-UI, so a display-only
component needs nothing added to the agent, on any framework.
Refs OSS-1034
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The add-menu ("+") button's tooltip hardcoded the string "Add attachments",
so `labels.chatInputToolbarAddButtonLabel` only retitled the menu item and
the tooltip stayed English. That blocked full localization of CopilotChat
without replacing the whole add-button slot.
Every other tooltip in the v2 chat surface is already label-driven, and the
Angular implementation already derives this tooltip from the same label, so
this was an oversight rather than a deliberate split.
The "/" shortcut glyph stays hardcoded — it is a key name, not prose.
Fixes#6750
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
## What does this PR do?
Adds `getLearningContainerId` to `CopilotKitIntelligence` so developers
can assign Intelligence Threads to Learning Containers without a
theta-prefixed Runtime option.
The selector receives:
- The resolved application `user`.
- The parsed AG-UI `input` for the run.
- The `agentId` and the `web` or `channel` surface.
Web runs pass the exact parsed `RunAgentInput`. Channel runs build the
same canonical input that the AgentRunner receives. Channels also carry
the resolved application user through the core and Intelligence adapter
boundaries.
The Runtime validates the selected stable ID and sends only that ID with
the existing Thread create or lock call. Intelligence stays responsible
for project scope, entitlements, Container lookup, and the one-time
Thread binding. Persisted AG-UI events remain the source for Learning
snapshots.
The old `ɵlearning` Runtime option remains as a deprecated fallback. The
Runtime rejects configurations that set both APIs.
## Why?
Learning Container assignment is an Intelligence SDK concern. Developers
also need the resolved user and complete run input to select a Container
from application data without reading raw transport details.
## Related PRs and Issues
- Refs
[ENT-1149](https://linear.app/copilotkit/issue/ENT-1149/enable-projects-to-learn-from-agent-runs-and-publish-reusable-skills)
- Related design: #6746
## Validation
- GitHub CI: 53 passed, 3 skipped
- `pnpm nx run-many -t test,check-types,build -p @copilotkit/runtime
@copilotkit/channels-core @copilotkit/channels-intelligence`
- `pnpm nx run-many -t publint,attw,check-dts -p @copilotkit/runtime
@copilotkit/channels-core @copilotkit/channels-intelligence`
- `pnpm lint` (0 errors; existing warnings remain)
- `npm run typecheck` and `npm run build` in `showcase/shell-docs`
- `npm test` in `showcase/shell-docs` has one pre-existing failure at
`inspector-docs.test.ts:141`: the tracked Threads callout contains
`Playground`.
## Checklist
- [x] I have read the [Contribution
Guide](https://github.com/copilotkit/copilotkit/blob/master/CONTRIBUTING.md)
- [x] If the PR changes or adds functionality, I have updated the
relevant documentation
- [x] Maintainer edits are available because this PR uses a branch in
the main repository
## The bug
`src/polyfills.ts` is five side-effect-only imports plus
`installStreamingFetch()`. The published barrel was 195 bytes:
```js
// node_modules/@copilotkit/react-native/dist/polyfills.mjs — 1.69.2
import { t as installStreamingFetch } from "./streaming-fetch-BnQh3vBz.mjs";
installStreamingFetch();
export { };
```
Every React Native app following the documented setup died on its first
runtime call:
```
E ReactNativeJS: '[CopilotKit] Error (runtime_info_fetch_failed):',
[ReferenceError: Property 'ReadableStream' doesn't exist]
```
## Root cause
The `sideEffects` field, but not in the way it first looks. It is
correct for *consumers* and wrong for *this package's own build*:
```json
"sideEffects": ["./dist/index.*", "./dist/headless.*", "./dist/polyfills.*", "./dist/polyfills/**/*"]
```
tsdown/rolldown reads the package's own `sideEffects` while bundling and
matches it against **source** paths. `src/polyfills/streams.ts` matches
none of those `dist` globs, so it is declared side-effect-free — a hard
assertion that lets rolldown drop the import without analysing the
`globalThis` assignments inside.
Reproduced in isolation at the pinned tsdown (0.20.3):
| `sideEffects` | built barrel |
|---|---|
| `["./dist/polyfills.*", "./dist/polyfills/**/*"]` | `export { };` —
empty |
| same + `["./src/polyfills.*", "./src/polyfills/**/*"]` | `import
"./polyfills/streams.mjs";` |
| field absent | `import "./polyfills/streams.mjs";` |
**Wider than the ticket recorded:** `dist/index.mjs` and
`dist/headless.mjs` also had zero polyfill code, so the package's
advertised auto-install on first import did not happen either. Not
RN-specific in principle — but I surveyed every package at `origin/main`
and this is the only one exposed. The other `sideEffects` arrays
(`react-core`, `react-ui`, `react-textarea`) are `["**/*.css"]`, which
matches source and works.
## The fix
Add matching `./src/**` globs. Barrel goes 195B → 362B with all five
imports; `headless.mjs` now leads with `import "./polyfills.mjs"`.
## The test, and why the existing one didn't catch this
`src/__tests__/polyfills.test.ts` has ~20 assertions covering all five
groups and was green the whole time — it imports `"../polyfills"`, the
TypeScript **source**, which vitest transpiles without bundling and
therefore without tree-shaking. It exercises a graph the published
package does not contain.
So the new check runs against `dist/`. Two things it has to get right to
be honest:
- **Node ships these globals natively.** Asserting `ReadableStream` is
"defined" after import passes on an empty barrel. The probe clears all
nine first, emulating Hermes.
- **The two formats need different treatment.** CJS is executed for real
in a child realm. ESM is checked structurally — it cannot be executed
here because `encoding.mjs` takes a named import from CommonJS
`text-encoding`, which Metro rewrites to a `require()` but bare Node ESM
rejects.
It is wired into `build`, so a dead barrel fails the build rather than
reaching npm — which matters, because this shipped through a fully green
suite.
## Docs
Added the `Property 'ReadableStream' doesn't exist` symptom to
troubleshooting, which previously covered only the inverse case (a
polyfill *conflict*).
I deliberately left the reference docs' "auto-installs on first import"
claims and the crypto import-order callout alone: both become **true**
once the build is fixed, and I verified the auto-install behaviourally.
## Verification
- **Red/green proven, not assumed:** reverted the `sideEffects` change,
rebuilt → 5/5 groups FAIL in both formats. Restored → 5/5 PASS. There is
also a test for a *single* group regressing, which a whole-barrel
assertion would wave through.
- **Packed tarball** (`pnpm pack`) verified behaviourally: all nine
globals install.
- 289 vitest + 26 script tests pass; `check-types` clean; `attw` green;
`publint` clean apart from a pre-existing `repository.url` suggestion;
oxfmt/oxlint clean.
- Added `{projectRoot}/scripts/**` to the package's `test` inputs and
confirmed cache invalidation (19/19 cached → 18/19 after touching the
verifier); without it, editing the verifier alone would restore a cached
pass.
**Not verified:** the on-device round trip — no emulator in this
environment. The bare-realm equivalent passes on the packed tarball.
## Follow-up worth its own ticket
`dist/polyfills/encoding.mjs` uses a named import from CommonJS
`text-encoding`. Metro handles it; a true-ESM consumer would not.
Pre-existing and not RN-facing, so left out of this change.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Closes OSS-1017, OSS-1018, OSS-1019.
Runtime volume can't currently be counted from the telemetry data alone.
Three separate reasons, all in the two `TelemetryClient`s, all fixed
here.
## What was wrong
**OSS-1017 — v2 samples but doesn't say so.**
`packages/runtime/src/v2/runtime/telemetry/telemetry-client.ts` gates
anonymous events at 5% and lets identified callers through at 100%, then
sends without stamping `sampleRate` / `sampleWeight`. The v1 client has
computed that block for a while; v2 never did. Measured in PostHog
(project 26816, `oss.runtime.copilot_request_created`), the share of
runtime volume arriving unweightable was 1.1% in May, 12.2% in June,
20.2% in July, 24.4% in August — roughly doubling each month. It has
already produced wrong GTM numbers: raw counts understate real volume
10–18× *and* overstate growth (+211% vs a true +128% Jan→Jul), because
the sampled/unsampled mix drifts as v2 adoption rises.
**OSS-1018 — identified v2 events aren't detectable downstream.** The
telemetry id travels as the `X-CopilotKit-Telemetry-Id` header, not an
event property, so whether an event was sampled at 5% or captured at
100% wasn't recoverable from the event. Consumers had to probe for
whatever identity fields happened to be present, which only ~2.7% of v2
events carry.
**OSS-1019 — v1 writes every event twice.** `capture()` sends to both
`lambdaClient.send()` and `segment.track()`, so one request becomes two
PostHog rows with nothing marking them as copies. Both prior dedupe
attempts produced wrong numbers: a "≥ 1.60 dual-emits" version cutoff is
wrong (1.59.5 dual-emits too — it added ~13M phantom requests to June),
and Segment-only drops the entire v2 runtime (19.5M in August).
## What changed
A new `packages/shared/src/telemetry/sampling.ts` holds
`computeSamplingMeta`, and both clients call it. The v1 client's inline
block is replaced by that call with identical output, so the
pre-existing v1 tests act as the regression check that the extraction is
behavior-preserving. v2 passes the result in `globalProperties`, the
slot the sink already spreads into the event and the same one v1 uses —
so both emitters are now identical on the wire.
`telemetry_identified` is carried explicitly rather than inferred from
`sampleWeight === 1`. Under `COPILOTKIT_TELEMETRY_SAMPLE_RATE=1`
anonymous events also weigh 1, and weight alone stops separating the
populations.
For the dual-write, per the issue's option 3 plus a dedupe key —
additive only, both transports keep flowing:
| property | value |
| -- | -- |
| `telemetry_emitter` | `v1-shared` \| `v2-runtime` |
| `telemetry_transport` | `segment` \| `lambda` |
| `telemetry_event_id` | one uuid per v1 `capture()`, identical on both
copies |
v2 doesn't carry an event id — it has a single transport, so there's
nothing to dedupe.
## The counting rule this enables
Drop v1's lambda copy, keep everything else, and sum the stamped weight:
```sql
sumIf(sampleWeight, NOT (telemetry_transport = 'lambda' AND telemetry_emitter = 'v1-shared'))
```
Note for whoever runs the next month-close: **this release breaks the
existing runbook query.** It isolates v2 with `NOT
JSONHas(properties,'sampleRate')`, which matches nothing once v2 starts
stamping `sampleRate` — `v2_runtime` would silently read zero and total
volume would lose ~24%. The runbook has been updated with an
era-agnostic query that gives the same answer either side of the
release; historical events don't get backfilled, so the old inference
path stays as the fallback for pre-fix data.
## Testing
Run from the worktree with `@copilotkit/shared` built so the runtime
resolves the real helper rather than a mock.
**`packages/shared` — 35 passed (35)**, including all 20 pre-existing v1
tests, unchanged. New `sampling.test.ts` (5) covers both branches, the
`sampleRate=1` collision that motivates the explicit flag, and an
empty-string telemetry id. New cases in `telemetry-client.test.ts` (4)
cover both copies sharing one `telemetry_event_id`, a fresh id per
capture, the emitter marker on both copies, and `telemetry_identified`
tracking the gate branch.
```
✓ src/telemetry/sampling.test.ts (5 tests)
✓ src/telemetry/lambda-client.test.ts (6 tests)
✓ src/telemetry/telemetry-client.test.ts (24 tests)
Test Files 3 passed (3)
Tests 35 passed (35)
```
**`packages/runtime` v2 telemetry — 37 passed (37)**, covering the new
client suite plus the four pre-existing license/telemetry integration
files:
```
✓ src/v2/runtime/telemetry/__tests__/telemetry-client.test.ts (6 tests)
✓ src/v2/runtime/telemetry/__tests__/global-properties.test.ts (6 tests)
✓ src/v2/runtime/telemetry/__tests__/instance-created.test.ts (5 tests)
✓ src/v2/runtime/__tests__/telemetry.test.ts (15 tests)
✓ src/v2/runtime/__tests__/sse-license-telemetry.integration.test.ts (1 test)
✓ src/v2/runtime/__tests__/sse-license-env-fallback.integration.test.ts (1 test)
✓ src/v2/runtime/__tests__/license-telemetry-endpoints.integration.test.ts (3 tests)
Tests 37 passed (37)
```
One pre-existing assertion changed: `global-properties.test.ts` asserted
`globalProperties` equals `{}` when the caller sets none, which is wrong
by design now that sampling metadata always rides there. Rewritten to
pin the exact key set, which preserves the original intent (nothing of
the caller's is added) and is strictly stronger.
**Mutation-checked** — each new test was verified to fail when its
mechanism is broken, then the mutation reverted and green confirmed:
| mutation | result |
| -- | -- |
| `effectiveSampleRate` always `= sampleRate` (drop identified branch) |
shared 2 failed, runtime 1 failed |
| `telemetry_identified` hardcoded `false` | shared 3 failed, runtime 1
failed |
| v1 event id hoisted per-client instead of per-capture | shared 1
failed |
| both v1 copies stamped `transport: "lambda"` | shared 1 failed |
| v2 drops the sampling block from `globalProperties` | runtime 4 failed
|
| *(reverted)* | shared 35/35, runtime 17/17 |
**Build/lint** — `tsdown` clean for both `@copilotkit/shared` (115
files) and `@copilotkit/runtime` (414 files); `oxfmt` applied; `oxlint`
reports 0 errors. The one warning on new code
(`consistent-function-scoping` on a test-local `jwtWith`) is the same
pattern the existing v1 test file already uses at
`telemetry-client.test.ts:42`.
Not run locally: the pre-commit `test-and-check-packages` hook, which
builds the whole monorepo and fails in this worktree on packages that
were never installed there (`core`, `sdk-js`, `channels-ui`) —
environmental, unrelated to these files. Leaving that to CI.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Moves the published packages from 0.0.57 to the current AG-UI release across
@ag-ui/client, core, encoder and proto — 27 declarations in 18 packages.
0.0.59 is the first release carrying the subagent protocol surface
(SUBAGENT_STARTED/FINISHED/ERROR, subagentRunId) along with the null-omission
cleanup, so this is the dependency CopilotKit's subagent work needs.
Scope is packages/** plus the release script noted below. The examples and
showcases sit on a spread of older pins (0.0.40 through 0.0.58) and are left
alone.
One behavioural change comes with the bump. channels-core ships
sanitizeAgentEventStream because @ag-ui/client used to reject a TOOL_CALL_START
carrying parentMessageId: null — the shape @ag-ui/langgraph emits for an
interrupt-triggering tool call. 0.0.59 accepts that null and treats it as
absent, so the two tests asserting the run dies WITHOUT the sanitizer no longer
hold. They now assert the run survives, and the one at agent level still checks
the tool call actually arrives so it cannot pass vacuously. The sanitizer is
untouched and its coercion tests are unchanged; it is simply no longer the
thing keeping such a run alive.
The bump also broke the packed Angular consumer matrix. That job generates a
smoke app from scripts/release/lib/angular-package.ts, whose manifest restated
"@ag-ui/client": "0.0.57" as a literal while packages/angular moved to 0.0.59.
pnpm then installed both copies and the app failed to compile:
TS2322: Type 'SmokeAgent' is not assignable to type 'AbstractAgent'.
Types have separate declarations of a private property '_debug'.
The smoke app imports AbstractAgent directly, so it has to resolve the identical
copy the library ships against. Read that version off the packed manifest --
which verify-angular-package.ts already parses for the Angular support contract
-- instead of restating it, so no future AG-UI bump can desynchronise it.
Inspector Threads now has **Try from here**. One click copies a stored
thread into a Playground scratch session. The stored thread does not
change.
If the copy fails, Inspector stays on Threads and keeps the current
Playground scratch. Example tour threads and locked Threads do not show
the button.
## What does this PR do?
Adds **Try from here** on a real stored thread in Inspector Threads. One
click copies messages and thread state into a Playground scratch
session. The stored thread does not change.
If the copy fails, Inspector stays on Threads and keeps the current
Playground scratch. Example tour threads and locked Threads do not show
the button.
## Related PRs and Issues
- Linear: OSS-873
- Playground base: https://github.com/CopilotKit/CopilotKit/pull/6580
(merged)
## Checklist
- [x] I have read the [Contribution
Guide](https://github.com/copilotkit/copilotkit/blob/master/CONTRIBUTING.md)
- [x] If the PR changes or adds functionality, I have updated the
relevant documentation
- [x] "Allow edits by maintainers" is checked
## Testing
**Commands run**
1. Rebased `feat/oss-873-try-from-here` onto `origin/main` and resolved
6 conflict files.
2. `npx nx run @copilotkit/web-inspector:test` — 626 tests passed (after
the stale-result guard).
3. `npx nx run @copilotkit/web-inspector:check-types` — passed.
**Manual test**
1. Open Inspector on localhost with Intelligence on, so a real stored
thread exists.
2. Open that thread. Confirm **Try from here** is in the thread header.
3. Click **Try from here**. Confirm Inspector opens Playground with the
copied messages and the stored thread is unchanged.
4. Open an example tour thread. Confirm **Try from here** is not shown.
5. Force a copy failure (disconnect runtime). Confirm Inspector stays on
Threads and the prior Playground scratch is unchanged.
**How this PR makes testing easy**
- `packages/web-inspector/src/__tests__/inspector-navigation.spec.ts`
covers the button, copy path, failure path, and a stale click that must
not overwrite Playground.
- `packages/web-inspector/src/lib/__tests__/telemetry.test.ts` covers
`oss.inspector.threads_try_from_here_clicked`.
## Risk / rollback
Risk is limited to Inspector Threads and Playground. A revert of this PR
removes the button and the new telemetry event. No runtime protocol
change.
## Public API change
New Inspector telemetry export and event name:
**Before**
```ts
trackThreadsTabClicked(props);
```
**After**
```ts
trackThreadsTabClicked(props);
trackThreadsTryFromHereClicked({ ...props, outcome: "success" });
```
`CpkThreadInspector` also emits a `tryFromHere` custom event when the
user clicks the button.
Closes OSS-904.
## The problem
`CopilotKitCore.runtimeConnectionStatus` was set only by the `/info`
handshake, which runs **once on connect**. If the runtime became
unreachable after that, the status stayed `connected` indefinitely —
measured two independent ways against `examples/v2/react/demo` and
recorded in the ticket.
The failure was not lost, it was filed in the wrong drawer: it arrived
as `agent_run_failed`, indistinguishable from an agent bug. So
everything downstream inherited the wrong answer — System Health
reported healthy, the launcher error signal could not fire for the most
common real symptom ("it worked a minute ago"), and a customer `onError`
handler written to separate wiring problems from agent problems got the
wrong classification.
## What this changes
The status now reports **the outcome of the last actual contact with the
runtime**.
A failed runtime request — or silence past a per-request watchdog —
triggers **one** bounded confirmation request. If nothing answers, the
status moves to `error` and the failure is emitted through the existing
wiring error code, so customers already handling startup wiring failures
pick up the mid-session case without changing a line. A subsequent
successful request re-syncs and clears it.
Crucially, **the conversation survives**. The transition does not
discard runtime knowledge, so the agent backing an open chat is the same
instance, its messages stay on screen, and submitting stays possible —
which matters because submitting is what restores the status.
No polling, no heartbeat, no retry loop. Every timer is bound to one
request and dies with it.
## Decisions worth knowing when reviewing
- **Reactive in both directions.** A heartbeat would put permanent
background traffic into every embedding application; a retry loop mostly
races a user who is about to retry anyway. The cost is stated rather
than hidden: while nothing is happening, nothing is detected.
- **Status change is separated from discarding knowledge.** The only
pre-existing code that set `error` also cleared `remoteAgents`. That is
right at startup and destructive mid-session, because conversation state
lives on the agent instance. Four sites now hold this invariant up
together; each carries a comment saying so.
- **The trigger is deliberately permissive, and the check is the
arbiter.** A request that received a successful response never triggers
a check; user cancellation never does; everything else may. Defining the
trigger precisely would mean maintaining a status-code list that is
complete only for the deployment topologies someone thought of.
- **Silence counts.** A server can refuse (fails fast) or hang (accepts
and never answers). A stopped dev server refuses; a container
mid-rollout, a half-switched deploy and a dropped tunnel hang. Only
bounding the check does not help, because no check starts — hence the
per-request watchdog. It observes only and never cancels the request.
- **The rule is stated by destination, not by call site**, so a runtime
route added later inherits the behaviour. Excluded: the Intelligence
realtime endpoint (a different service — reporting its outage as
"runtime unreachable" would be a false diagnosis), endpoints belonging
to the customer, and the stop request.
- **Recovery may prune, under two conditions**: the runtime must have
reported at least one agent, and the agent must carry no conversation
state. An empty list is the signature of a runtime that has not finished
registering.
- **"Answered but refused" keeps the error status and gets a different
message.** An expired token means the app cannot work, so red is right;
telling the reader "unreachable" would send them to check ports and
containers.
## Deliberately not delivered
- Detecting an outage, or a recovery, while the application is idle.
- Recovery by opening the Threads view: every binding withholds thread
requests until the status is already connected, so nothing is sent while
it is red. The thread plumbing still earns its place for *detection*.
- A signal for the Intelligence realtime endpoint failing while the
runtime is healthy — a real gap, and its own ticket.
- Memory and suggestion routes adopting the instrumented fetch.
- A new status value or a new error code.
## Costs this introduces
`error` now means two things — "never connected, no agents" and "lost
mid-session, agents intact". Documented on the enum. And because the
status can now change mid-session at all, an outage costs some churn
that did not exist before: the memory list and the Inspector's thread
list are cleared and refetched, and where the chat owns its run-activity
store it is stopped and restarted. All of it is paid on a user-caused
transition, never while idle.
## Testing
Four independent reviewers audited an earlier revision of this branch;
the ten defects they reproduced are fixed and each is pinned by a test
that was red first. A mutation audit of 110 mutants killed 90; the
surviving holes were closed in the round after.
The connection-health suites carry 72 tests. Request counting is a
first-class assertion throughout, because several decisions are
*absences* — no polling, no retry loop, one check per burst, no traffic
while red — and an absence is only testable by counting. Those tests use
fake timers advancing ten minutes; that boundary is documented where it
lives, since anything slower is invisible to them.
Verified by hand in a browser with the runtime running as its own
process, so the page outlives it: a refusing runtime, a hanging runtime,
recovery, an agent added during an outage, an agent deleted during an
outage. `performance.timeOrigin` was checked throughout to prove the
page never reloaded and the result was not an artefact of a fresh
handshake.
## Follow-ups this leaves behind
Three of these deserve their own ticket. None blocks this PR; all three
are consequences of where its scope was drawn, and they are listed here
so the boundary is explicit rather than implied.
### 1. A signal for the Intelligence realtime endpoint
In Intelligence mode the browser gets its chat events from a **second
service** at its own address; the runtime is only asked for the
credentials. If that service fails while the runtime is healthy, this
change correctly reports the runtime as reachable — and the user
experiences exactly the silence this ticket exists to remove.
It is excluded here on purpose: folding it into the runtime status would
report "runtime unreachable" about a healthy runtime, and a false
diagnosis costs more debugging time than no signal. It needs its own
signal, which is a presentation decision as much as a detection one.
### 2. Memory routes onto the instrumented fetch
The memory store still builds with the global fetch, so its
runtime-bound requests are invisible to connection health. Two costs: a
genuine failure there is a signal we discard, and a success there cannot
restore the status.
The asymmetry is what makes this worth fixing rather than leaving:
memory is the surface most disrupted by a status transition (its list is
cleared and refetched) and currently the one least able to contribute.
The change itself is small — that module already takes its request
function as an injected dependency.
### 3. Consumers should key on what they need, not on the status value
Several consumers treat "status is not connected" as "discard
everything": the memory list, the Inspector's thread list, and the
chat's run-activity store. That was harmless while the status could not
change after page load. It can now, so every outage costs churn that did
not exist before.
This is the same mistake this PR fixes three times *inside* core — a
guard bound to a state instead of to the thing it protects. The
principle was applied internally and not to these consumers. That makes
the churn listed under "Costs" above **deferred rather than inherent**,
and it is the largest of the three follow-ups: three consumers in three
packages, each with its own risk, which is why it was kept out of this
PR.
### Two smaller items
- The launcher error signal on `main` carries a comment stating the
limitation this change removes ("a runtime that dies after the page
loaded … raises nothing … closing that gap means a re-probe in the
core"). It becomes false when this lands and should be corrected then.
- `packages/web-inspector/src/styles/generated.css` is build output
under version control and re-dirties the tree on every build. Unrelated
to this PR, but the Tailwind source glob scans test files, so any prose
comment containing a utility word (`fixed`, `hidden`, `visible`,
`block`) silently changes the committed CSS. Narrowing the glob would
remove the class of problem.
Full specification, including the interview decisions and every
revision: `OSS-904-PRD.md`.
## Why
The Inspector said what Intelligence *is* and linked out to a signup
page. Of ~1,655 Inspector opens in 90 days, **under 100 clicked any
CTA**. This replaces the feature list with an argument, and the outbound
link with an install that happens in the editor the developer is already
in.
This is the unfinished half of OSS-867, whose body asks for exactly
this: *"If a capability requires Intelligence, detail why and include a
video demonstrating that feature working end-to-end."*
## What changed
**A four-slide argument, paired to the picture beside it.** Each slide
carries two sentences and the visual they describe: your users' threads
→ the pattern inside them → the skill file → what it compounds into. An
earlier draft sold Threads in prose while animating Learning; bound
together they read as one chain, and `meeting-scheduling/SKILL.md`
recurs through all four so the closing diagram is checkable rather than
decorative.
Condensed from the six-phase animation on the Intelligence home page —
not screen-recorded. A ported version is themeable, stays sharp, and
costs no asset weight; the original also runs 21.4s and opens on the
agent booking the wrong meeting, a poor first frame for a card arguing
for the product.
**A copy-prompt button instead of a link out.** It hands the CLI's own
onboarding prompt to a coding agent. Every previous Intelligence CTA
opened a new tab into a signup form, which is where developers drop out.
It carries the CLI's `onboarding_run_id`, so
`oss.inspector.home_prompt_copied` can be joined to
`cli.onboarding.completed` on the Intelligence side. `home_cta_clicked`
only ever proved that someone clicked a link — this is the first event
that can show whether an install followed.
**Section anatomy mirrors System Health** (header band, rule, content),
so the action sits in the same top-right slot the status pill and renew
link already use, and the panel keeps one section shape throughout.
## Correctness of the claims
The copy was checked against the product's own surfaces, and two claims
did not survive:
- **Skills are not applied at run time.** Candidates land at
`pending_review`, a human approves, and the published set is pulled down
with `copilotkit skills download`. Nothing reads published skills during
a run. The slide says approve → pull in → the next run starts from what
worked.
- **Insights were missing**, and with them the evidence link that makes
Learning credible: every Insight cites the Threads and messages behind
it.
Also: *Rich Threads* is the product's name for the durable ones, and the
distinction is the whole sale next to a Threads tab full of local ones
that die on reload. "Your users" means the app's end users — which is
what the platform means too (`identifyUser` resolves one user per
request; a thread carries `end_user_id`, renamed from `user_id` because
the old name *"caused repeated misdiagnosis"*).
## Behaviour worth reviewing
- The story advances **only while Home is visible and the document is
not hidden**. A debugging tool should not hold a repeating timer behind
a closed panel.
- Slide motion is horizontal and derived from each slide's index
relative to the active one, so clicking a tab backwards animates
backwards with no stored direction to fall out of sync.
- Copied state **expires after 4s** so the button invites a second
press; a failed copy **does not**, because that state is the only place
the prompt is selectable by hand.
- Three modes, not two: a lapsed plan keeps the renew link and never
sees an install prompt.
- The rotating copy is hidden from assistive tech (it would announce
four times a loop); one stable sentence is exposed in its place and is
test-covered so it cannot quietly rot.
## Deliberate omissions
**No third-party coding-agent logos on the button**, unlike the
Intelligence app. That app is a private hosted surface; this package
ships inside other people's sites, and vendoring Anthropic's and
OpenAI's marks is not a call to make quietly. The helper line names the
agents in text.
Deferred and worth discussing separately: ordering the Home sections by
state (health first when something is broken, Intelligence first when
nothing is), and putting the same button in the locked Learning and
Threads tabs, where intent is highest.
## Verification
617 tests pass, `check-types` clean, oxlint 0 errors. Verified live in
both themes: all four slides, uniform 16px padding on every slide, card
height stable across slides, copy success **and** failure paths, and the
header band unchanged at 76px when the copied hint appears. The reset
behaviour is mutation-checked — the file records which mutation each
test does and does not catch.
Five packages publish to npm with no `license` field, so registry
metadata and automated license scanners report them as **Unknown**:
```
@copilotkit/agentcore-runner published=1.68.1 license=<NONE>
@copilotkit/core published=1.68.1 license=<NONE>
@copilotkit/sqlite-runner published=1.68.1 license=<NONE>
@copilotkit/voice published=1.68.1 license=<NONE>
@copilotkit/web-inspector published=1.68.1 license=<NONE>
```
The repo is MIT (see `LICENSE`) and every other published
`@copilotkit/*` package already declares it — these five were simply
missed. This adds `"license": "MIT"` to each, positioned before
`"repository"` to match the sibling packages.
## Why
Reported downstream in #2860, where a corporate procurement scan refused
packages whose license it could not resolve. That class of scanner reads
the `license` field from registry metadata; a `LICENSE` file in the repo
is not enough, and these packages ship no `LICENSE` file either.
**Correcting the record on that issue while I am here:** the `@ag-ui/*`
packages named in the original report are *not* affected. Every version
the reporter’s scanner flagged already carries `"license": "MIT"`:
```
@ag-ui/client@0.0.42 MIT
@ag-ui/core@0.0.37 MIT
@ag-ui/core@0.0.42 MIT
@ag-ui/encoder@0.0.42 MIT
@ag-ui/langgraph@0.0.20 MIT
@ag-ui/proto@0.0.42 MIT
```
`@ag-ui/core` has declared MIT since at least 0.0.35. An earlier triage
note on #2860 attributed the failure to a missing SPDX field upstream;
that was wrong, and why their scanner reported `Unknown` for `@ag-ui/*`
is still unexplained. This PR fixes the part that is genuinely defective
on our side.
## Testing
Metadata-only; no source, build, or runtime change.
- Confirmed the five missing fields against the live registry with `npm
view <pkg> license` (output above), and confirmed the other published
`@copilotkit/*` packages (`runtime`, `react-core`, `react-ui`, `shared`,
`sdk-js`, `angular`, `channels`, `channels-core`) already report `MIT`.
- Enumerated every non-private `packages/*/package.json` on
`origin/main` to confirm these five are the complete set missing the
field.
- Each edited file re-parsed with `json.load` and reports `MIT`.
- The `sync-lockfile` pre-commit hook resolved all 71 workspace projects
against the edited manifests without error.
Placement matches `packages/shared/package.json`, where `"license"`
immediately precedes `"repository"`.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
The assertion searched footer.outerHTML for "241" to prove the unclamped
thread count never reaches the user. outerHTML also carries lit's part
markers, which lit builds as `lit$` + nine digits from Math.random(),
regenerated per process. Roughly one process in a hundred rolls a marker
containing those digits and fails the assertion with no relation to what the
footer rendered — this run drew lit$924125892$.
Comments are stripped before the check. Visible text and attributes still
count, so a genuine leak in an aria-label is caught exactly as before, and the
stripping is asserted so it cannot silently stop working.
Not introduced here: the same dice roll could hit any change to this package.
The other outerHTML assertions in the suite compare two strings from the same
process and share a marker, so they were never exposed.
The rail's four tabs reported nothing. Adds
oss.inspector.home_story_beat_selected, carrying the step as a property rather
than one event per label: the labels are expected to move as the story is
iterated, and per-label events would retire with them. beat_index rides along
so a reorder can be judged against where people actually click.
Only a press reports. The story also advances on its own every few seconds,
and reporting that would emit one event per idle developer per beat and bury
the handful of real interactions under a metronome — asserted, not assumed.
The v2 runtime client gated anonymous events at 5% and let identified
callers through at 100%, then sent without recording which branch the
event took. A quarter of runtime volume — 24.4% in August and roughly
doubling each month — arrived carrying no record of its own sampling, so
it could not be weighted from the data alone. Downstream had to hardcode
a x20 assumption, which both understates real volume and overstates
growth as the sampled/unsampled mix drifts.
Extract the v1 client's sampling block into shared/telemetry/sampling so
the two clients cannot drift again, and call it from both. Identified
events weigh 1, anonymous ones 1/sampleRate.
Carry telemetry_identified explicitly rather than letting consumers infer
identity from sampleWeight === 1: under COPILOTKIT_TELEMETRY_SAMPLE_RATE=1
anonymous events also weigh 1 and the two populations stop being
distinguishable.
The v1 client also sends every capture to both Segment and the lambda
sink, so one request produces two rows with nothing marking them as
copies. Stamp telemetry_emitter, telemetry_transport, and a per-capture
telemetry_event_id shared by both copies, making the dedupe explicit
instead of inferred from $lib. Both transports keep flowing.
Refs OSS-1017, OSS-1018, OSS-1019
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
`src/polyfills.ts` is five side-effect-only imports. The `sideEffects` globs only matched
`./dist/**`, and rolldown matches that field against SOURCE paths while bundling, so every
`src/polyfills/*.ts` was declared pure and dropped. Every published version through 1.69.2
shipped a 195-byte barrel installing nothing but streaming fetch, so an app following the
documented setup died on its first runtime call with `Property 'ReadableStream' doesn't exist`.
`dist/index.mjs` and `dist/headless.mjs` lost the same imports, so the package's advertised
auto-install on first import did not happen either.
Add matching `./src/**` globs. The barrel goes 195B to 362B with all five imports, and
`headless.mjs` now leads with `import "./polyfills.mjs"`.
`src/__tests__/polyfills.test.ts` stayed green throughout this, because it imports the source,
which is never bundled and so is never tree-shaken. Add `scripts/verify-polyfill-barrel.mjs`,
which checks `dist/` instead. It clears the nine globals first (Node ships them natively and
Hermes does not, so asserting they are merely "defined" would pass on an empty barrel), then
executes the CJS barrel in a child realm and checks the ESM barrel structurally. ESM cannot be
executed here: the encoding polyfill takes a named import from CommonJS `text-encoding`, which
Metro rewrites to a require() but bare Node ESM rejects.
The check runs from `build`, so a dead barrel fails the build rather than reaching npm.
Also document the `ReadableStream doesn't exist` symptom in troubleshooting, where only the
inverse case (a polyfill *conflict*) was covered before.
Verified: reverting the sideEffects change and rebuilding turns the check red in both formats,
5 of 5 groups; restoring it turns it green. A behavioural check on the packed tarball installs
all nine globals. 289 vitest + 26 script tests pass.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The spec replaced @copilotkit/core wholesale with a two-export factory. That
held until the Inspector started mounting in these tests — it is enabled by
default in browser frameworks now, and its connectedCallback calls
isInspectorThreadBridgeEnabled, one of seventeen value exports it imports from
core. A missing one throws an uncaught exception, so the run fails while all
49 test files still report passing, which is a confusing way to find out.
The factory now spreads the real module and overrides only CopilotKitCore and
the connection-status enum, which is what these tests actually drive. Listing
the seventeen would have postponed the next occurrence rather than removed it.
Surfaced by the web-inspector work on this branch: angular only runs when
affected, and it becomes affected the moment web-inspector changes — so the
first PR to touch web-inspector after the default-on change was going to hit
this regardless of what it changed.
The checked-in artifact was produced mid-rebase, so its Tailwind token block
reflects the utilities in use at that point rather than on today's main. A
fresh build:css on the rebased tree differs; this is that output, so the
committed file matches what the build produces.
"What Intelligence does" promised an explanation and pointed at
intelligence.copilotkit.ai — the product and signup page. Mis-promising a
destination is a poor trade at the moment the card is asking to be trusted,
and the label was doing no work besides.
The two actions are two routes to one outcome: let the coding agent wire it
up, or go and do it in the browser. The secondary now says so — "Set it up
yourself" — which is honest about where it lands and quietly argues for the
primary, because "yourself" implies the other route is not.
The confirmation used to stand for the rest of the session. The reasoning was
that a developer leaves for their editor and comes back and should still find
the instruction — but whoever comes back has already pasted. The likelier
reason to return is to copy again, and a button wearing a checkmark and
"Prompt copied" reads as spent even though it still works.
Copied now reverts after 4s: button label and secondary line both return, so
the "What Intelligence does" link comes back too. Longer than the 2s the
Threads setup prompt uses, because this state carries an instruction to read
and not just an acknowledgement.
A failed copy does not expire. That state is the only place the prompt text is
selectable by hand, and pulling it away mid-selection would be worse than the
clipboard failing.
Three tests, mutation-checked: disabling the reset kills the two copied-state
tests. The failed-state test is blunter and only fails when both safeguards go
(the scheduling condition and the timer's own state re-check) — recorded in the
file so nobody reads more into it than it proves.
The instruction was a second row under the button, which pushed the action
column past the band's 76px min-height and shoved the whole story down at the
moment the developer had just acted — the worst possible time for the layout
to move.
It now takes over the secondary slot instead of adding to it. Before the press
the useful aside is "what is this"; after it, "where to put it". Both are
single lines in the same flex slot, so the swap cannot change the height:
measured 76px and an unmoved story in both states.
Dropping the line was the alternative, but it is the one instruction that
cannot go: a prompt on the clipboard with no idea what to do with it converts
nobody. It is also vendor-neutral now — naming three editors was brittle and
the prompt identifies the agent itself.
- "Your users had thousands" carried a count a developer wiring this up
locally does not have yet. "Your users have all the others" holds on day one
and at scale. Whose users: the app's end users, which is what the platform
means too — `identifyUser` resolves one user per request from the app, and a
thread carries `end_user_id`, a column renamed away from `user_id` because
the old name "caused repeated misdiagnosis" against control-plane users.
- The Skills line defended instead of selling. "Nothing reaches your agent
until you approve it" answers a fear the reader has not voiced and plants
the worry it deflects. Same fact as ownership: a SKILL.md that is yours to
review, edit and ship.
- The last tab is "Intelligence", not "Better agents". The other three are
the parts; this one is the whole, so the rail reads Threads · Learning ·
Skills · Intelligence. Beat id renamed to match, selectors included.
Checked the four slides against the product's own copy instead of against
intuition, and two claims did not survive.
- "Skills apply it — without you writing another prompt" said the platform
applies skills at run time. It does not: candidates land at pending_review,
a human approves, and the published set is pulled down with
`copilotkit skills download`. Nothing reads published skills during a run.
The slide now says approve, pull in, and the next run starts from what
worked — which is also the stronger pitch, because a developer does not
want a platform silently changing how their agent behaves.
- Insights were missing entirely, and with them the evidence link that makes
Learning credible: every Insight cites the Threads and messages behind it.
Learning's own onboarding leads with "46 evidence refs" for that reason.
Also: "Rich Threads" is the product's name for the durable ones, and the
distinction is the whole sale next to a Threads tab full of local ones that
die on reload. A skill is a directory holding a SKILL.md, so the card shows
`meeting-scheduling/SKILL.md` and is marked Pending review.
- The rail's last step was "Reuse", which named neither a product surface nor
an outcome. It is "Better agents" — the promise, in the customer's words.
- Secondary link before the primary button: with the filled button in the
middle it read as a block wedged between the heading and the link instead
of the one thing to press.
- The heading gets the brand mark and its band a wash along the brand's hue
path, reusing the account strip's existing device. Not gradient text: at
18px it renders muddy, costs contrast, and is the most over-used premium
tell going.
The Inspector said what Intelligence is and linked out to a signup page. Of
~1,655 opens in 90 days, under 100 clicked any CTA. This replaces the feature
list with an argument, and the outbound link with an install that happens in
the editor the developer is already in.
- Copy: the heading names the product; a four-slide argument runs underneath,
each slide pairing two sentences with the picture beside it (threads →
the pattern in them → the skill file → it applies itself). An earlier draft
sold Threads in prose while animating Learning; bound together they read as
one chain.
- Copy prompt: hands the CLI's own onboarding prompt to a coding agent instead
of opening a signup form. Carries the CLI's onboarding_run_id, so
oss.inspector.home_prompt_copied can finally be joined to
cli.onboarding.completed — home_cta_clicked only ever proved a click.
A refused clipboard reveals the prompt instead of swallowing it.
- Three modes, not two: a lapsed plan keeps the renew link and never sees an
install prompt.
- Section anatomy mirrors System Health (header band, rule, content), so the
action sits in the same top-right slot the status pill and renew link use.
- The story only advances while Home is visible and the document is not
hidden; a debugging tool should not hold a repeating timer behind a closed
panel.
No third-party coding-agent logos on the button, unlike the Intelligence app:
this package ships inside other people's sites, and vendoring those marks is
not a call to make quietly. The helper line names the agents in text.
Rendering activity messages outside of CopilotChatMessageView (custom chat
shells, dashboards, headless setups) currently requires instantiating the
whole message view or copying its private resolution logic. Tool calls
already have a public component for this (RenderToolCalls), activities do
not.
This adds CopilotActivity (<copilot-activity [message] [agentId]>), a
standalone host for a single activity message. CopilotChatMessageView now
delegates to it, so there is one implementation for the activity role. No
behavior change. The renderer resolution moves into an internal
pickActivityRenderer that is not exported.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The oxlint CI job failed on the ext-apps migration:
- copilotkit/no-single-arg-zod-record (error): the ui/message _meta field used
the single-arg z.record(z.any()), which is a compile-time error against Zod 4.
Use the two-arg form z.record(z.string(), z.any()).
- Remove the dead hand-rolled JSON-RPC message types left over from before the
AppBridge migration (no longer referenced).
Also normalize type-only imports and formatting in the touched e2e tests.
Yann Jouanin